[
https://issues.apache.org/jira/browse/HDFS-2136?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13065418#comment-13065418
]
Matt Foley commented on HDFS-2136:
----------------------------------
Just to be very concrete, here is one of the FI test cases I would like to see
implemented, and don't see how to do well without AOP:
Start the Namenode with 4 Storage Directories with pre-existing namespace data.
During startup with upgrade, it reads all 4 SD's, then writes to all 4 SD's.
Test scenario: The NN shall successfully read both FSImage and Edits from all 4
SD's. However, during the write phase it shall experience (injected) failure
to write the new Edits log to the second SD. Expected result: the NN shall
detect the failure and respond by exiting, per HDFS-1955.
> 1073: Fault injection for StorageDirectory failures during read/write of
> FSImage/Edits files
> --------------------------------------------------------------------------------------------
>
> Key: HDFS-2136
> URL: https://issues.apache.org/jira/browse/HDFS-2136
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Matt Foley
>
> Both HDFS-1955 and HDFS-2135 have observed that it is difficult to unit test
> such failures. As a result, regression of HDFS-1955 was only found by
> careful manual review (thanks, atm!). Since 1073 is making broad changes to
> the way these files are read and written, and appropriately putting effort
> into correct error handling, I propose we make also make it possible to
> auto-test that error handling.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
